build_from_source_en.md 7.7 KB
Newer Older
L
liaogang 已提交
1
Installing from Sources
G
gangliao 已提交
2
==========================
Z
zhangjinchao01 已提交
3

L
liaogang 已提交
4 5 6
* [1. Download and Setup](#download)
* [2. Requirements](#requirements)
* [3. Build on Ubuntu](#ubuntu)
Q
qijun 已提交
7 8
* [4. Build on Centos](#centos)

L
liaogang 已提交
9

L
liaogang 已提交
10
## <span id="download">Download and Setup</span> 
G
gangliao 已提交
11
You can download PaddlePaddle from the [github source](https://github.com/PaddlePaddle/Paddle).
Z
zhangjinchao01 已提交
12

L
liaogang 已提交
13
```bash
G
gangliao 已提交
14
git clone https://github.com/PaddlePaddle/Paddle paddle
15
cd paddle
L
liaogang 已提交
16 17
```
## <span id="requirements">Requirements</span>
Z
zhangjinchao01 已提交
18

L
liaogang 已提交
19 20 21
To compile the source code, your computer must be equipped with the following dependencies.

- **Compiler**: GCC >= 4.8 or Clang >= 3.3 (AppleClang >= 5.1)
L
liaogang 已提交
22
- **CMake**: version >= 3.0 (at least CMake 3.4 on Mac OS X)
L
liaogang 已提交
23
- **BLAS**: MKL, OpenBlas or ATLAS
Z
zhangjinchao01 已提交
24

L
liaogang 已提交
25 26 27
**Note:** For CUDA 7.0 and CUDA 7.5, GCC 5.0 and up are not supported!
For CUDA 8.0, GCC versions later than 5.3 are not supported!

L
liaogang 已提交
28
### Options
Z
zhangjinchao01 已提交
29

L
liaogang 已提交
30
PaddlePaddle supports some build options. 
Z
zhangjinchao01 已提交
31

G
gangliao 已提交
32 33 34 35 36 37 38 39 40
<html>
<table> 
<thead>
<tr>
<th scope="col" class="left">Optional</th>
<th scope="col" class="left">Description</th>
</tr>
</thead>
<tbody>
L
liaogang 已提交
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55
<tr><td class="left">WITH_GPU</td><td class="left">Compile PaddlePaddle with NVIDIA GPU</td></tr>
<tr><td class="left">WITH_AVX</td><td class="left">Compile PaddlePaddle with AVX intrinsics</td></tr>
<tr><td class="left">WITH_DSO</td><td class="left">Compile PaddlePaddle with dynamic linked CUDA</td></tr>
<tr><td class="left">WITH_TESTING</td><td class="left">Compile PaddlePaddle with unit testing</td></tr>
<tr><td class="left">WITH_SWIG_PY</td><td class="left">Compile PaddlePaddle with inference api</td></tr>
<tr><td class="left">WITH_STYLE_CHECK</td><td class="left">Compile PaddlePaddle with style check</td></tr>
<tr><td class="left">WITH_PYTHON</td><td class="left">Compile PaddlePaddle with python interpreter</td></tr>
<tr><td class="left">WITH_DOUBLE</td><td class="left">Compile PaddlePaddle with double precision</td></tr>
<tr><td class="left">WITH_RDMA</td><td class="left">Compile PaddlePaddle with RDMA support</td></tr>
<tr><td class="left">WITH_TIMER</td><td class="left">Compile PaddlePaddle with stats timer</td></tr>
<tr><td class="left">WITH_PROFILER</td><td class="left">Compile PaddlePaddle with GPU profiler</td></tr>
<tr><td class="left">WITH_DOC</td><td class="left">Compile PaddlePaddle with documentation</td></tr>
<tr><td class="left">ON_COVERALLS</td><td class="left">Compile PaddlePaddle with code coverage</td></tr>
<tr><td class="left">COVERALLS_UPLOAD</td><td class="left">Package code coverage data to coveralls</td></tr>
<tr><td class="left">ON_TRAVIS</td><td class="left">Exclude special unit test on Travis CI</td></tr>
G
gangliao 已提交
56
</tbody>
G
gangliao 已提交
57
</table>
G
gangliao 已提交
58
</html>
Z
zhangjinchao01 已提交
59

L
liaogang 已提交
60
**Note:**
L
liaogang 已提交
61 62
  - The GPU version works best with Cuda Toolkit 8.0 and cuDNN v5.
  - Other versions like Cuda Toolkit 7.0, 7.5 and cuDNN v3, v4 are also supported.
L
liaogang 已提交
63
  - **To utilize cuDNN v5, Cuda Toolkit 7.5 is prerequisite and vice versa.**
Z
zhangjinchao01 已提交
64

L
liaogang 已提交
65
As a simple example, consider the following:  
Z
zhangjinchao01 已提交
66

L
liaogang 已提交
67
1. **BLAS Dependencies(optional)**
Z
zhangjinchao01 已提交
68
  
L
liaogang 已提交
69 70
    CMake will search BLAS libraries from system. If not found, OpenBLAS will be downloaded, built and installed automatically.
    To utilize preinstalled BLAS, you can simply specify MKL, OpenBLAS or ATLAS via `MKL_ROOT`, `OPENBLAS_ROOT` or `ATLAS_ROOT`.
Z
zhangjinchao01 已提交
71

L
liaogang 已提交
72
    ```bash
L
liaogang 已提交
73 74 75 76
    # specify MKL
    cmake .. -DMKL_ROOT=<mkl_path>
    # or specify OpenBLAS
    cmake .. -DOPENBLAS_ROOT=<openblas_path>
L
liaogang 已提交
77
    ```
Z
zhangjinchao01 已提交
78

L
liaogang 已提交
79
2. **Doc Dependencies(optional)**
Z
zhangjinchao01 已提交
80

L
liaogang 已提交
81
    To generate PaddlePaddle's documentation, install dependencies and set `-DWITH_DOC=ON` as follows:
Z
zhangjinchao01 已提交
82

L
liaogang 已提交
83 84
    ```bash
    pip install 'sphinx>=1.4.0'
85
    pip install sphinx_rtd_theme recommonmark
Z
zhangjinchao01 已提交
86

L
liaogang 已提交
87 88 89 90
    # install doxygen on Ubuntu
    sudo apt-get install doxygen 
    # install doxygen on Mac OS X
    brew install doxygen
Z
zhangjinchao01 已提交
91

L
liaogang 已提交
92 93 94
    # active docs in cmake
    cmake .. -DWITH_DOC=ON`
    ```
Z
zhangjinchao01 已提交
95

L
liaogang 已提交
96
## <span id="ubuntu">Build on Ubuntu 14.04</span>
Z
zhangjinchao01 已提交
97

L
liaogang 已提交
98
### Install Dependencies
Z
zhangjinchao01 已提交
99

L
liaogang 已提交
100
- **CPU Dependencies**
Z
zhangjinchao01 已提交
101

L
liaogang 已提交
102 103 104
    ```bash
    # necessary
    sudo apt-get update
L
liaogang 已提交
105
    sudo apt-get install -y g++ make cmake build-essential python python-pip libpython-dev git
L
liaogang 已提交
106 107
    sudo pip install wheel numpy
    sudo pip install 'protobuf>=3.0.0'
L
liaogang 已提交
108 109 110
    ```
  
- **GPU Dependencies (optional)**
Z
zhangjinchao01 已提交
111

L
liaogang 已提交
112
    To build GPU version, you will need the following installed:
Z
zhangjinchao01 已提交
113

L
liaogang 已提交
114 115 116 117
        1. a CUDA-capable GPU
        2. A supported version of Linux with a gcc compiler and toolchain
        3. NVIDIA CUDA Toolkit (available at http://developer.nvidia.com/cuda-downloads)
        4. NVIDIA cuDNN Library (availabel at https://developer.nvidia.com/cudnn)
Z
zhangjinchao01 已提交
118

L
liaogang 已提交
119 120 121 122 123
    The CUDA development environment relies on tight integration with the host development environment,
    including the host compiler and C runtime libraries, and is therefore only supported on
    distribution versions that have been qualified for this CUDA Toolkit release.
        
    After downloading cuDNN library, issue the following commands:
Z
zhangjinchao01 已提交
124

L
liaogang 已提交
125 126 127 128
    ```bash
    sudo tar -xzf cudnn-7.5-linux-x64-v5.1.tgz -C /usr/local
    sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
    ```
129
    Then you need to set LD\_LIBRARY\_PATH, PATH environment variables in ~/.bashrc.
L
liaogang 已提交
130 131 132 133 134 135 136 137 138

    ```bash
    export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
    export PATH=/usr/local/cuda/bin:$PATH
    ```

### Build and Install

As usual, the best option is to create build folder under paddle project directory.
Z
zhangjinchao01 已提交
139 140

```bash
L
liaogang 已提交
141
mkdir build && cd build
L
liaogang 已提交
142
``` 
L
liaogang 已提交
143

L
liaogang 已提交
144
Finally, you can build and install PaddlePaddle:
Z
zhangjinchao01 已提交
145 146 147

```bash
# you can add build option here, such as:    
L
liaogang 已提交
148
cmake .. -DCMAKE_INSTALL_PREFIX=<path to install>
149
# please use sudo make install, if you want to install PaddlePaddle into the system
Z
zhangjinchao01 已提交
150
make -j `nproc` && make install
L
liaogang 已提交
151
# set PaddlePaddle installation path in ~/.bashrc
L
liaogang 已提交
152
export PATH=<path to install>/bin:$PATH
L
liaogang 已提交
153
# install PaddlePaddle Python modules.
L
liaogang 已提交
154
sudo pip install <path to install>/opt/paddle/share/wheels/*.whl
Z
zhangjinchao01 已提交
155
```
Q
qijun 已提交
156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
## <span id="centos">Build on Centos 7</span>

### Install Dependencies

- **CPU Dependencies**

    ```bash
    # necessary
    sudo yum update
    sudo yum install -y epel-release
    sudo yum install -y make cmake3 python-devel python-pip gcc-gfortran swig git
    sudo pip install wheel numpy
    sudo pip install 'protobuf>=3.0.0'
    ```
  
- **GPU Dependencies (optional)**

    To build GPU version, you will need the following installed:

        1. a CUDA-capable GPU
        2. A supported version of Linux with a gcc compiler and toolchain
        3. NVIDIA CUDA Toolkit (available at http://developer.nvidia.com/cuda-downloads)
        4. NVIDIA cuDNN Library (availabel at https://developer.nvidia.com/cudnn)

    The CUDA development environment relies on tight integration with the host development environment,
    including the host compiler and C runtime libraries, and is therefore only supported on
    distribution versions that have been qualified for this CUDA Toolkit release.
        
    After downloading cuDNN library, issue the following commands:

    ```bash
    sudo tar -xzf cudnn-7.5-linux-x64-v5.1.tgz -C /usr/local
    sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
    ```
    Then you need to set LD\_LIBRARY\_PATH, PATH environment variables in ~/.bashrc.

    ```bash
    export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
    export PATH=/usr/local/cuda/bin:$PATH
    ```

### Build and Install

As usual, the best option is to create build folder under paddle project directory.

```bash
mkdir build && cd build
``` 

Finally, you can build and install PaddlePaddle:

```bash
# you can add build option here, such as:    
cmake3 .. -DCMAKE_INSTALL_PREFIX=<path to install>
# please use sudo make install, if you want to install PaddlePaddle into the system
make -j `nproc` && make install
# set PaddlePaddle installation path in ~/.bashrc
export PATH=<path to install>/bin:$PATH
# install PaddlePaddle Python modules.
sudo pip install <path to install>/opt/paddle/share/wheels/*.whl
```